We propose to develop a brain tumor center at the University of Cincinnati, building on existing clinical strengths and expertise in the basic sciences. As documented, our neurosurgery program sees a large number of patients with a broad spectrum of brain tumors. The program plans to draw upon and integrate basic neuro- oncology-related research at the Institution with ongoing clinical activities. To this end, we have submitted three basic research proposals that utilize clinical materials and/or expertise immediately or in the near future. First, Dr. Anil Menon wishes to expand on his observations that chromosome aberrations on the long arm of chromosome 14 occur frequently in malignant meningioma, and the suggestion that LOH on 14q represents a late event in the transition from benign to malignant meningioma. He proposes also to utilize cytogenetic and molecular genetic approaches to characterize variously graded meningiomas to "allelotype" meningiomas of different malignancy grades. Ultimately this will lead to cloning of suppressor genes related to meningioma. The second project, directed by Dr. Robert Brackenbury, utilizes a somatic cell genetic approach to identify, characterize and eventually isolate tumor suppressor genes that may contribute to the genesis and progression of glioblastomas. The approach entails the defining of regulatory regions within two genes that are not expressed in normal glial cells but are expressed in glioblastomas, consistent with loss of suppressor function. These regulatory elements will be linked to selectable markers that afford forward and reverse selection. By transfecting selectable genes into somatic cell hybrids that suppress or permit expression and by subsequent transfection of cDNAs that alter expression of the selectable genes, it should be possible to recover and identify cDNAs and genes with suppressor activity. Lastly, Dr. Parysek will direct a project to develop replication defective retroviral vectors containing genes with potential cytopathic capacity to target medulloblastomas in vitro and in vivo. In one approach, the cytopathic gene will be an HSVtk gene, which will render infected cells sensitive to ganciclovir, a common anti-herpetic drug. In a second approach, the cytopathic gene will be an intermediate filament disrupter gene whose expression will be toxic to infected cells. Since non-dividing cells are immune to retroviral infection, post-mitotic neurons will not become infected while proliferating tumor cells will. Retrovirus will be applied by direct injection to subcutaneous tumors and by stereotactic methods to intracranial tumors in nude mice. Establishment of a brain tumor center will directly benefit the proposed basic science programs (and others) and the clinical investigators by providing a formal vehicle for their coordination and integration. The basic scientists provide many cellular and molecular tools for developing new diagnostics and treatments, while clinical investigator involvement can indicate how these tools can be best utilized and can accelerate their implementation.